word_counter 0.1.0 → 0.2.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 5c147e993b823171dd15fde073ea9b756e90f9f2
-  data.tar.gz: ac1b7049265125eecc4e8bc2ba0667aa240480be
+  metadata.gz: 369c75efe673d26c303a3a337ec9ac52374f224a
+  data.tar.gz: 8851cd6e31789b9ffce17a9e202ed168a8e6ffcc
 SHA512:
-  metadata.gz: 08f066c3857a101ef68d26ff4d83a80ca0cf3857a38761abc5c9b862debddd31d0771293ef4b08a80ac84a540254f125c5c3792aaa0030dfa159645ae3c00ebd
-  data.tar.gz: 4dffa4cf7ff6ccc62df4b4181ff142555b780fbe5a02dc175a29358790bbefdfca99800ce0023a9efff271f76c7a29b3c51a854c6a5546b55a555c305313dbbc
+  metadata.gz: e4a8b6d8b6347e2389f9da6f2d380094663f65e57a6e364fd7c5b52fb2c08ef4595202ff04c879426a9886e23bd562d46701ba8da432993e4059406bb45522ae
+  data.tar.gz: 8b808a9111151250694eb470ff8472cc9cf5203c101b090e76eeb6dfbe3f4f601b5c6a48a32b42f7c9b946822f3ed48670c71ae1d8ab9e4bb78d7fe12415ac77

data/.travis.yml

@@ -0,0 +1,7 @@
+language: ruby
+rvm:
+  - "1.9.3"
+  - "2.0.0"
+  - "2.1.0"
+  - "2.1.1"
+script: bundle exec rspec spec

data/README.md

@@ -2,6 +2,8 @@
 
 Counts words from either a file or a website, and prints a report to stdout.
 
+[![Build Status](https://api.travis-ci.org/wulftone/word_counter.svg?branch=master)](http://travis-ci.org/wulftone/word_counter)
+
 ## Installation
 
 Add this line to your application's Gemfile:

data/lib/word_counter/version.rb

@@ -1,3 +1,3 @@
 class WordCounter
-  VERSION = "0.1.0"
+  VERSION = "0.2.0"
 end

data/lib/word_counter.rb

@@ -1,61 +1,137 @@
 require "word_counter/version"
-require 'curb'
+require "net/http"
 require 'nokogiri'
 
 class NoFileError < StandardError; end
+class NoWebsiteError < StandardError; end
 
 class WordCounter
 
 
   ##
+  # WordCounter!
+  #
   # @param filename [String] The path and filename of the file to analyze
+  # @param show_sentences [Boolean] (default: false) If true, WordCounter will print out the sentences which contain the counted word in question
   def initialize arg, show_sentences = false
     raise ArgumentError, "Please supply a URL or file path." unless arg
     @show_sentences = true if show_sentences
 
     begin
       # try to open it as a file
-      analyze_file arg
+      @hashified_words = WordCounter.analyze_file arg
     rescue NoFileError => e
       # try to analyze it as a website, so curl it
-      analyze_website arg
+      @hashified_words = WordCounter.analyze_website arg
+    end
+  end
+
+
+  ##
+  # Helper method
+  def show_sentences?
+    @show_sentences
+  end
+
+
+  ##
+  # Prints a report to stdout
+  def report
+    hashified_words_with_sorted_lines = @hashified_words.each do |word, data|
+      # data[:lines].sort
+    end
+
+    sorted_hash = hashified_words_with_sorted_lines.sort_by { |word, data|
+      [-data[:count], word]
+    }
+
+    sorted_hash.each do |word, data|
+      puts "#{data[:count]} #{word}"
+      puts "    #{data[:lines].join("\n    ")}" if show_sentences?
     end
   end
 
 
-  def analyze_website url
-    http = Curl.get url
-    html = Nokogiri::HTML http.body_str
+  ##
+  # Fetch a url
+  #
+  # @param uri_str [String] A URI
+  def self.fetch(uri_str, limit = 10)
+    raise ArgumentError, 'too many HTTP redirects' if limit == 0
+
+    uri = URI uri_str
+    response = Net::HTTP.get_response uri
+
+    case response
+    when Net::HTTPSuccess then
+      response
+    when Net::HTTPRedirection then
+      location = response['location']
+      warn "redirected to #{location}"
+      fetch(location, limit - 1)
+    else
+      response.value
+    end
+  end
+
+
+  ##
+  # Prepends an http:// if there isn't one.
+  #
+  # @param arg [String]
+  def self.urlize arg
+    if arg =~ /^(http:\/\/|https:\/\/)/
+      arg
+    else
+      "http://#{arg}"
+    end
+  end
+
 
-    html.search('script').remove
-    html.search('meta').remove
-    html.search('style').remove
-    text = html.text
-    @hashified_words = WordCounter.hashify_words text
+  ##
+  # Vists a website and analyzes it
+  #
+  # @param arg [String] A website URL
+  def self.analyze_website arg
+    url = WordCounter.urlize arg
+    res = WordCounter.fetch url
+    raise NoWebsiteError unless res.code == '200'
+
+    doc = Nokogiri::HTML res.body
+    doc.search('script').remove
+    doc.search('meta').remove
+    doc.search('style').remove
+    text = doc.text
+    hashify_words text
   end
 
 
 
-  def analyze_file filename
-    raise NoFileError, "File does not exist!" unless File.exist? filename
+  ##
+  # Opens a file and analyzes it
+  #
+  # @param file [String] A path to a file
+  def self.analyze_file file
+    raise NoFileError, "File does not exist!" unless File.exist? file
+
+    hashified_words = nil
 
-    @file = File.open filename do |file|
-      @hashified_words = WordCounter.hashify_words file
+    @file = File.open file do |file|
+      hashified_words = hashify_words file
     end
+
+    hashified_words
   end
 
   ##
   # Builds the data structures we use for our analysis.
   #
-  # @param file [File] The file we're analyzing
-  def self.hashify_words file
+  # @param string [File] The string we're analyzing (notice: can also be a File object, because `each_line` also works with Files.)
+  def self.hashify_words string
     hash = {}
 
-    file.each_line do |line|
-      # words = line.split
-      words = line.split(/\W+/)
-
-      words.reject! { |w| w.empty? }
+    string.each_line do |line|
+      words = line.split(/\W+/).reject { |w| w.empty? }
 
       words.each do |word|
         sym = word.to_sym
@@ -75,27 +151,4 @@ class WordCounter
 
     hash
   end
-
-
-  def show_sentences?
-    @show_sentences
-  end
-
-
-  ##
-  # Prints a report to stdout
-  def report
-    hashified_words_with_sorted_lines = @hashified_words.each do |word, data|
-      data[:lines].sort
-    end
-
-    sorted_hash = hashified_words_with_sorted_lines.sort_by { |word, data|
-      [-data[:count], word]
-    }
-
-    sorted_hash.each do |word, data|
-      puts "#{data[:count]} #{word}"
-      puts "    #{data[:lines].join("\n    ")}" if show_sentences?
-    end
-  end
 end

data/spec/word_counter_spec.rb

@@ -60,13 +60,6 @@ describe WordCounter do
   end
 
 
-  it '.hashify_words' do
-    File.open test_file do |file|
-    result = WordCounter.hashify_words file
-      result.should eq hash
-    end
-  end
-
   it '#analyze_website' do
     wc = WordCounter.new 'www.example.com'
 
@@ -102,4 +95,16 @@ describe WordCounter do
 1 without
 "
   end
+
+
+  it '.urlize' do
+    url = WordCounter.urlize 'example.com'
+    url.should eq 'http://example.com'
+    url = WordCounter.urlize 'https://example.com'
+    url.should eq 'https://example.com'
+    url = WordCounter.urlize 'http://example.com'
+    url.should eq 'http://example.com'
+  end
+
+
 end

data/word_counter.gemspec

@@ -24,6 +24,5 @@ Gem::Specification.new do |spec|
   spec.add_development_dependency "guard-rspec"
   spec.add_development_dependency "pry-debugger"
 
-  spec.add_dependency "curb"
   spec.add_dependency "nokogiri"
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: word_counter
 version: !ruby/object:Gem::Version
-  version: 0.1.0
+  version: 0.2.0
 platform: ruby
 authors:
 - trevor bortins
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2014-03-25 00:00:00.000000000 Z
+date: 2014-03-26 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: bundler
@@ -80,20 +80,6 @@ dependencies:
     - - ">="
       - !ruby/object:Gem::Version
         version: '0'
-- !ruby/object:Gem::Dependency
-  name: curb
-  requirement: !ruby/object:Gem::Requirement
-    requirements:
-    - - ">="
-      - !ruby/object:Gem::Version
-        version: '0'
-  type: :runtime
-  prerelease: false
-  version_requirements: !ruby/object:Gem::Requirement
-    requirements:
-    - - ">="
-      - !ruby/object:Gem::Version
-        version: '0'
 - !ruby/object:Gem::Dependency
   name: nokogiri
   requirement: !ruby/object:Gem::Requirement
@@ -118,6 +104,7 @@ extra_rdoc_files: []
 files:
 - ".gitignore"
 - ".rspec"
+- ".travis.yml"
 - Gemfile
 - Guardfile
 - LICENSE.txt